Kingston DataTraveler and Kingston IronKey USB flash drives provide on-the-go file storage for photos, music, video and more. They are available in both standard and encrypted security for home, school, office and enterprise organisations.
If hostd restarts during a long-running quiesced snapshot operation, hostd might automatically run a snapshot consolidation to remove redundant disks and improve the virtual machine performance. However, the consolidation operation might race with the running quiesced snapshot operation and the virtual machines stop responding.
Flash Disk Lock 1.7 ~UPD~
The P2MCache lock is a spin lock compromising the fairness across execution threads that wait for the P2MCache. As a result, the ESXi host might fail with purple diagnostic screen and the following error: 2017-09-15T06:50:17.777Z cpu11:153493)@BlueScreen: Spin count exceeded - possible deadlock with PCPU 19 2017-09-15T06:50:17.777Z cpu11:153493)Code start: 0x418034400000 VMK uptime: 8:21:18:05.527 2017-09-15T06:50:17.778Z cpu11:153493)Saved backtrace from: pcpu 19 SpinLock spin out NMI 2017-09-15T06:50:17.778Z cpu11:153493)0x439127a1bb60:[0x4180345911bf]VmMemPin_ReleasePhysMemRange@vmkernel#nover+0x1f stack: 0x0 2017-09-15T06:50:17.778Z cpu11:153493)0x439127a1bbb0:[0x41803455d706]P2MCache_Release@vmkernel#nover+0xa6 stack: 0x4393bb7a7000 2017-09-15T06:50:17.779Z cpu11:153493)0x439127a1bbf0:[0x4180345658a2]PhysMem_ReleaseSGE@vmkernel#nover+0x16 stack: 0x0
When you use devices with a physical block size other than 512 or 4096 bytes, in the /var/log/vmkernel.log file of the ESXi host you might see multiple warning messages similar to: ScsiPath: 4395: The Physical block size "8192" reported by the path vmhba3:C0:T1:L0 is not supported.
Inefficient block allocation mechanisms might lead to multiple iterations of metadata reads. This causes a long quiescing time while creating and extending thick provisioned lazy zeroed VMDK files on a VMFS6 datastore.
This issue is specific to vSphere Virtual Volumes datastores when a VMDK file is assigned to different SCSI targets across snapshots. The lock file of the VMDK is reassigned across different snapshots and might be incorrectly deleted when you revert the virtual machine to a snapshot. Due to the missing lock file, the disk does not open, and the virtual machine fails to power on.
If you have a virtual machine on a SeSparse snapshot and you query the physical layout of the VMDK from the Guest Operating System or a third-party application, a physical CPU lockup might be triggered if the VMDK file size is not a multiple of 4K. As a result, the ESXi host fails with a purple screen.
If you migrate a virtual machine with a NetX filter across clusters by using vSphere vMotion, the virtual machine might lose network connectivity. As a side effect of service profile optimization, the NetX filter might get blocked after the migration. As a result, the virtual machine loses network connectivity.
After a host fails, the Entry Persistence Daemon (EPD) might not restart. Some components continue to persist after an object is deleted. Messages similar to the following appear in the EPD log: 2018-07-10T10:35:21.029Z 71308 -- XXX: Lock file '/scratch/epd-store.db' exists. 2018-07-10T10:35:21.029Z 71308 -- XXX: Did the host or EPD recently crash? 2018-07-10T10:35:21.029Z 71308 -- XXX: Assuming it's OK. Unlinking lock file.. 2018-07-10T10:35:21.030Z 71308 Failed to delete lock file: Is a directory
After an interruption of a live migration involving one or more disks, such as a Storage vMotion or a vMotion without shared storage, the guest operating system might become unresponsive for up to 8 minutes, due to a rare race condition.
Virtual machines might go into invalid state and you cannot complete tasks such as power on, create, delete, migrate and reconfigure, depending on which locks are stuck in transactions as a result of an abrupt reboot of the ESXi host.
When you use devices with a physical block size other than 512 or 4096 Bytes, in the /var/log/vmkernel.log file of the ESXi host you might see multiple warning messages similar to: ScsiPath: 4395: The Physical block size "8192" reported by the path vmhba3:C0:T1:L0 is not supported.
Inefficient block allocation mechanisms might lead to multiple iterations of metadata reads. This causes a long quiesce time while creating and extending thick provisioned lazy zeroed VMDK files on a VMFS6 datastore. 2ff7e9595c
Comments